
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.model_selection import train_test_split
from sklearn.tree import DecisionTreeClassifier, plot_tree
from sklearn.metrics import classification_report, confusion_matrix, accuracy_score, silhouette_score
from sklearn.cluster import KMeans
from sklearn.preprocessing import StandardScaler
from sklearn.decomposition import PCA
from sklearn.linear_model import LinearRegression
from sklearn.metrics import ConfusionMatrixDisplay, precision_score, recall_score, f1_score, r2_score
from sklearn.neighbors import KNeighborsClassifier
from sklearn.datasets import make_classification
from ydata_profiling import ProfileReport
import plotly.express as px
import plotly.figure_factory as ff
import plotly.graph_objects as go


filename = 'DataSet-Troconiz-E1.csv'
url = "DataSet-Troconiz-E1.csv"
dataset = pd.read_csv(url)


pd.set_option('display.max_columns', 50)
dataset.head()


dataset.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 7043 entries, 0 to 7042
Data columns (total 46 columns):
 #   Column                             Non-Null Count  Dtype  
---  ------                             --------------  -----  
 0   Customer ID                        7043 non-null   object 
 1   Referred a Friend                  7043 non-null   object 
 2   Number of Referrals                7043 non-null   int64  
 3   Tenure in Months                   7043 non-null   int64  
 4   Offer                              7043 non-null   object 
 5   Phone Service                      7043 non-null   object 
 6   Avg Monthly Long Distance Charges  7043 non-null   float64
 7   Multiple Lines                     7043 non-null   object 
 8   Internet Service                   7043 non-null   object 
 9   Internet Type                      7043 non-null   object 
 10  Avg Monthly GB Download            7043 non-null   int64  
 11  Online Security                    7043 non-null   object 
 12  Online Backup                      7043 non-null   object 
 13  Device Protection Plan             7043 non-null   object 
 14  Premium Tech Support               7043 non-null   object 
 15  Streaming TV                       7043 non-null   object 
 16  Streaming Movies                   7043 non-null   object 
 17  Streaming Music                    7043 non-null   object 
 18  Unlimited Data                     7043 non-null   object 
 19  Contract                           7043 non-null   object 
 20  Paperless Billing                  7043 non-null   object 
 21  Payment Method                     7043 non-null   object 
 22  Monthly Charge                     7043 non-null   float64
 23  Total Regular Charges              7043 non-null   float64
 24  Total Refunds                      7043 non-null   float64
 25  Total Extra Data Charges           7043 non-null   float64
 26  Total Long Distance Charges        7043 non-null   float64
 27  Gender                             7043 non-null   object 
 28  Age                                7043 non-null   int64  
 29  Under 30                           7043 non-null   object 
 30  Senior Citizen                     7043 non-null   object 
 31  Married                            7043 non-null   object 
 32  Dependents                         7043 non-null   object 
 33  Number of Dependents               7043 non-null   int64  
 34  City                               7043 non-null   object 
 35  Zip Code                           7043 non-null   int64  
 36  Latitude                           7043 non-null   float64
 37  Longitude                          7043 non-null   float64
 38  Population                         7043 non-null   int64  
 39  Churn Value                        7043 non-null   int64  
 40  CLTV                               7043 non-null   int64  
 41  Churn Category                     1869 non-null   object 
 42  Churn Reason                       1869 non-null   object 
 43  Total Customer Svc Requests        7043 non-null   int64  
 44  Product/Service Issues Reported    7043 non-null   int64  
 45  Customer Satisfaction              1834 non-null   float64
dtypes: float64(9), int64(11), object(26)
memory usage: 2.5+ MB


dataset.describe()


# Calcular la frecuencia relativa
freq_rel = dataset['Churn Value'].value_counts(normalize=True)
# Convertir a DataFrame y resetear el índice
df_freq_rel = freq_rel.to_frame().reset_index()
df_freq_rel.columns = ['Churn Value', 'Frecuencia Relativa']
fig = px.bar(df_freq_rel, 
             x='Churn Value', 
             y='Frecuencia Relativa', 
             title='Distribución de la Tasa de Cancelación',
             labels={'Churn Value':'Valor de Churn, 0 no abandono, 1 abandono',
                     'Frecuencia Relativa':'Porcentaje'}, 
             color='Churn Value')

fig.update_yaxes(tickformat=".0%")  # Formato de porcentaje en el eje y
fig.update_xaxes(tickvals=[0, 1])
fig.show()


fig = px.histogram(dataset, 
                   x="Customer Satisfaction", 
                   nbins=5,  # Número de barras. 
                   title='Distribución de la Satisfacción del Cliente',
                   labels={'Customer Satisfaction':'Satisfacción del Cliente, mínimo 1, máximo 5', 'count':'Frecuencia'}, 
                   color='Customer Satisfaction')  # Color de las barras. En este caso, se está usando la misma variable de Customer Satisfaction.

fig.show()


conteo_satisfaccion = dataset['Customer Satisfaction'].value_counts().sort_index()
conteo_satisfaccion = conteo_satisfaccion.astype(int)
total_encuestas = conteo_satisfaccion.sum()
porcentaje_satisfaccion = round((conteo_satisfaccion / total_encuestas) * 100, 2)
df_satisfaccion = pd.DataFrame({'Conteo': conteo_satisfaccion, 'Porcentaje (%)': porcentaje_satisfaccion})
df_satisfaccion.loc['Total'] = [total_encuestas, 100.00]
df_satisfaccion.loc['Total', 'Conteo'] = df_satisfaccion.loc['Total', 'Conteo'].astype(int)
df_satisfaccion


fig = px.box(dataset, x='Customer Satisfaction', y='Tenure in Months', 
             labels={
                     "Customer Satisfaction": "Satisfacción del Cliente, minimo 1, maxima 5",
                     "Tenure in Months": "Tiempo de Permanencia (Meses)"
                 },
             title='Satisfacción del Cliente vs. Tiempo de Permanencia (Boxplot)')

fig.show()


import plotly.express as px

fig = px.box(dataset, y="Customer Satisfaction", 
              title='Boxplot de la Satisfacción del Cliente', 
              labels={"Customer Satisfaction": "Satisfacción del Cliente"})
fig.show()


fig = px.box(dataset, x='Churn Value', y='Customer Satisfaction', 
             labels={
                     "Churn Value": "Tasa de Cancelación, 0 no abandono, 1 abandono",
                     "Customer Satisfaction": "Satisfacción del Cliente, minimo 1, maxima 5"
                 },
             title='Tasa de Cancelación vs. Satisfacción del Cliente (Boxplot)')

fig.show()


mean_tenure = dataset.groupby('Churn Value')['Tenure in Months'].mean().reset_index()

fig = px.line(mean_tenure, x='Churn Value', y='Tenure in Months',
              labels={
                  "Churn Value": "Tasa de Cancelación, 0 no abandono, 1 abandono",
                  "Tenure in Months": "Tiempo de Permanencia Promedio (Meses)"
              },
              title='Tasa de Cancelación vs. Tiempo de Permanencia Promedio')

fig.show()


fig = px.box(dataset, x='Customer Satisfaction', y='Monthly Charge', 
             labels={
                     "Customer Satisfaction": "Satisfacción del Cliente, minimo 1, maxima 5",
                     "Monthly Charge": "Cargos Mensuales"
                 },
             title='Satisfacción del Cliente vs. Cargos Mensuales (Boxplot)')

fig.show()


import plotly.express as px

fig = px.strip(dataset, x='Churn Value', y='Monthly Charge',
               labels={
                     "Churn Value": "Tasa de Cancelación, 0 no abandono, 1 abandono",
                     "Monthly Charge": "Cargos Mensuales"
                 },
                 title='Tasa de Cancelación vs. Cargos Mensuales (Strip Plot)')

fig.show()


fig = px.box(dataset, x='Customer Satisfaction', y='Total Customer Svc Requests', 
             labels={
                     "Customer Satisfaction": "Satisfacción del Cliente, minimo 1, maxima 5",
                     "Total Customer Svc Requests": "Solicitudes de Servicio al Cliente"
                 },
             title='Satisfacción del Cliente vs. Solicitudes de Servicio al Cliente (Boxplot)')

fig.show()


import plotly.express as px

# Calcular la media de 'Total Customer Svc Requests' para cada 'Churn Value'
mean_requests = dataset.groupby('Churn Value')['Total Customer Svc Requests'].mean().reset_index()

fig = px.line(mean_requests, x='Churn Value', y='Total Customer Svc Requests',
              labels={
                  "Churn Value": "Tasa de Cancelación, 0 no abandono, 1 abandono",
                  "Total Customer Svc Requests": "Solicitudes de Servicio al Cliente"
              },
              title='Tasa de Cancelación vs. Solicitudes de Servicio al Cliente (Line Plot)')

fig.show()


fig = px.box(dataset, x='Customer Satisfaction', y='Product/Service Issues Reported', 
             labels={
                     "Customer Satisfaction": "Satisfacción del Cliente, minimo 1, maxima 5",
                     "Product/Service Issues Reported": "Problemas de Producto/Servicio Reportados"
                 },
             title='Satisfacción del Cliente vs. Problemas de Producto/Servicio Reportados (Boxplot)')

fig.show()


fig = px.box(dataset, x='Churn Value', y='Product/Service Issues Reported', 
             labels={
                     "Churn Value": "Tasa de Cancelación, 0 no abandono, 1 abandono",
                     "Product/Service Issues Reported": "Problemas de Producto/Servicio Reportados"
                 },
             title='Tasa de Cancelación vs. Problemas de Producto/Servicio Reportados (Boxplot)')

fig.show()


variables_interes = ['Churn Value','Tenure in Months', 'Avg Monthly Long Distance Charges', 'Avg Monthly GB Download',
                     'Monthly Charge', 'Total Regular Charges', 'Total Refunds', 'Total Extra Data Charges',
                     'Total Long Distance Charges', 'Age', 'Total Customer Svc Requests',
                     'Product/Service Issues Reported', 'Customer Satisfaction']

# Filtrar el dataframe con las variables seleccionadas
dataset_filtrado = dataset[variables_interes]

# Calcular la matriz de correlación
correlation_matrix = dataset_filtrado.corr()

# Redondear los valores de correlación a 2 decimales
correlation_matrix_rounded = correlation_matrix.round(2)

# Crear el mapa de calor utilizando Plotly
fig = ff.create_annotated_heatmap(z=correlation_matrix_rounded.values,
                                  x=correlation_matrix_rounded.columns.tolist(),
                                  y=correlation_matrix_rounded.columns.tolist(),
                                  colorscale='Viridis')
fig.show()


# Lista de columnas a eliminar
columns_to_drop = ['City', 'Zip Code', 'Latitude', 'Longitude', 'Population', 'Gender', 'Age', 'Under 30', 'Customer ID']

# Elimina las columnas
dataset = dataset.drop(columns=columns_to_drop)


import requests
base_url = 'https://randomuser.me/api/'
params = {'inc': 'gender,nat,phone,email,location,name,id', 'results': 5000}

new_dataset = pd.DataFrame()

#iterar 2 veces para obtener los 7043 registros 
#se le agrega un while para solo llenar los 7043
while len(new_dataset) < 7043:
    response = requests.get(base_url, params=params)
    data = response.json()
    api_df = pd.json_normalize(data['results'])

    # Concatena los datos de la API con tu DataFrame existente
    new_dataset = pd.concat([new_dataset, api_df], ignore_index=True)

    # Si la longitud del DataFrame supera 7043, se trunca a 7043
    if len(new_dataset) > 7043:
        new_dataset = new_dataset.iloc[:7043]


new_dataset.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 7043 entries, 0 to 7042
Data columns (total 19 columns):
 #   Column                          Non-Null Count  Dtype 
---  ------                          --------------  ----- 
 0   gender                          7043 non-null   object
 1   email                           7043 non-null   object
 2   phone                           7043 non-null   object
 3   nat                             7043 non-null   object
 4   name.title                      7043 non-null   object
 5   name.first                      7043 non-null   object
 6   name.last                       7043 non-null   object
 7   location.street.number          7043 non-null   int64 
 8   location.street.name            7043 non-null   object
 9   location.city                   7043 non-null   object
 10  location.state                  7043 non-null   object
 11  location.country                7043 non-null   object
 12  location.postcode               7043 non-null   object
 13  location.coordinates.latitude   7043 non-null   object
 14  location.coordinates.longitude  7043 non-null   object
 15  location.timezone.offset        7043 non-null   object
 16  location.timezone.description   7043 non-null   object
 17  id.name                         7043 non-null   object
 18  id.value                        5716 non-null   object
dtypes: int64(1), object(18)
memory usage: 1.0+ MB


duplicated_ids = new_dataset["id.value"].duplicated()
print("Number of duplicated IDs:", duplicated_ids.sum())

Number of duplicated IDs: 1377


new_dataset = new_dataset.drop_duplicates(subset='id.value')


new_dataset.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 5666 entries, 0 to 7042
Data columns (total 19 columns):
 #   Column                          Non-Null Count  Dtype 
---  ------                          --------------  ----- 
 0   gender                          5666 non-null   object
 1   email                           5666 non-null   object
 2   phone                           5666 non-null   object
 3   nat                             5666 non-null   object
 4   name.title                      5666 non-null   object
 5   name.first                      5666 non-null   object
 6   name.last                       5666 non-null   object
 7   location.street.number          5666 non-null   int64 
 8   location.street.name            5666 non-null   object
 9   location.city                   5666 non-null   object
 10  location.state                  5666 non-null   object
 11  location.country                5666 non-null   object
 12  location.postcode               5666 non-null   object
 13  location.coordinates.latitude   5666 non-null   object
 14  location.coordinates.longitude  5666 non-null   object
 15  location.timezone.offset        5666 non-null   object
 16  location.timezone.description   5666 non-null   object
 17  id.name                         5666 non-null   object
 18  id.value                        5665 non-null   object
dtypes: int64(1), object(18)
memory usage: 885.3+ KB


dataset['index'] = range(1, len(dataset) + 1)
new_dataset['index'] = range(1, len(new_dataset) + 1)


merged_df = pd.merge(dataset, new_dataset, on='index')


merged_df.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 5666 entries, 0 to 5665
Data columns (total 57 columns):
 #   Column                             Non-Null Count  Dtype  
---  ------                             --------------  -----  
 0   Referred a Friend                  5666 non-null   object 
 1   Number of Referrals                5666 non-null   int64  
 2   Tenure in Months                   5666 non-null   int64  
 3   Offer                              5666 non-null   object 
 4   Phone Service                      5666 non-null   object 
 5   Avg Monthly Long Distance Charges  5666 non-null   float64
 6   Multiple Lines                     5666 non-null   object 
 7   Internet Service                   5666 non-null   object 
 8   Internet Type                      5666 non-null   object 
 9   Avg Monthly GB Download            5666 non-null   int64  
 10  Online Security                    5666 non-null   object 
 11  Online Backup                      5666 non-null   object 
 12  Device Protection Plan             5666 non-null   object 
 13  Premium Tech Support               5666 non-null   object 
 14  Streaming TV                       5666 non-null   object 
 15  Streaming Movies                   5666 non-null   object 
 16  Streaming Music                    5666 non-null   object 
 17  Unlimited Data                     5666 non-null   object 
 18  Contract                           5666 non-null   object 
 19  Paperless Billing                  5666 non-null   object 
 20  Payment Method                     5666 non-null   object 
 21  Monthly Charge                     5666 non-null   float64
 22  Total Regular Charges              5666 non-null   float64
 23  Total Refunds                      5666 non-null   float64
 24  Total Extra Data Charges           5666 non-null   float64
 25  Total Long Distance Charges        5666 non-null   float64
 26  Senior Citizen                     5666 non-null   object 
 27  Married                            5666 non-null   object 
 28  Dependents                         5666 non-null   object 
 29  Number of Dependents               5666 non-null   int64  
 30  Churn Value                        5666 non-null   int64  
 31  CLTV                               5666 non-null   int64  
 32  Churn Category                     1869 non-null   object 
 33  Churn Reason                       1869 non-null   object 
 34  Total Customer Svc Requests        5666 non-null   int64  
 35  Product/Service Issues Reported    5666 non-null   int64  
 36  Customer Satisfaction              1509 non-null   float64
 37  index                              5666 non-null   int64  
 38  gender                             5666 non-null   object 
 39  email                              5666 non-null   object 
 40  phone                              5666 non-null   object 
 41  nat                                5666 non-null   object 
 42  name.title                         5666 non-null   object 
 43  name.first                         5666 non-null   object 
 44  name.last                          5666 non-null   object 
 45  location.street.number             5666 non-null   int64  
 46  location.street.name               5666 non-null   object 
 47  location.city                      5666 non-null   object 
 48  location.state                     5666 non-null   object 
 49  location.country                   5666 non-null   object 
 50  location.postcode                  5666 non-null   object 
 51  location.coordinates.latitude      5666 non-null   object 
 52  location.coordinates.longitude     5666 non-null   object 
 53  location.timezone.offset           5666 non-null   object 
 54  location.timezone.description      5666 non-null   object 
 55  id.name                            5666 non-null   object 
 56  id.value                           5665 non-null   object 
dtypes: float64(7), int64(10), object(40)
memory usage: 2.5+ MB


# Selecciona las características numéricas relevantes y la variable objetivo
feature_columns = ['Tenure in Months', 'Monthly Charge', 'Total Regular Charges', 'Total Refunds', 'Total Extra Data Charges', 'Total Long Distance Charges']
X = dataset[feature_columns]
y = dataset['Churn Value']

# Divide el conjunto de datos en conjuntos de entrenamiento y prueba
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# Entrena un árbol de decisión
clf = DecisionTreeClassifier(random_state=42)
clf.fit(X_train, y_train)

# Realiza predicciones en el conjunto de prueba
y_pred = clf.predict(X_test)

# Evalúa el modelo
print("Confusion Matrix:")
print(confusion_matrix(y_test, y_pred))
print("\nClassification Report:")
print(classification_report(y_test, y_pred))
print("Accuracy Score:", accuracy_score(y_test, y_pred))

Confusion Matrix:
[[858 151]
 [141 259]]

Classification Report:
              precision    recall  f1-score   support

           0       0.86      0.85      0.85      1009
           1       0.63      0.65      0.64       400

    accuracy                           0.79      1409
   macro avg       0.75      0.75      0.75      1409
weighted avg       0.79      0.79      0.79      1409

Accuracy Score: 0.7927608232789212


# Crear el clasificador de árbol de decisión
clf = DecisionTreeClassifier()

# Entrenar el clasificador con los datos de entrenamiento
clf.fit(X_train, y_train)

# Acceder a las características de importancia
importances = clf.feature_importances_

# Imprimir las características de importancia
print(importances)

[0.19435232 0.38196614 0.16615866 0.01074149 0.10858585 0.13819554]


# Selecciona las características numéricas relevantes y la variable objetivo
feature_columns = ['Tenure in Months', 'Monthly Charge', 'Total Regular Charges', 'Total Refunds', 'Total Extra Data Charges', 'Total Long Distance Charges']

importances = clf.feature_importances_  # Asumiendo que clf es tu clasificador y tiene los valores de importancia

# Ordena las columnas de características y las importancias en función de las importancias
sorted_columns, sorted_importances = zip(*sorted(zip(feature_columns, importances), key=lambda x: x[1], reverse=True))

# Crea el gráfico de barras horizontales
fig = go.Figure(data=[go.Bar(x=sorted_importances, y=sorted_columns, orientation='h')])

# Personaliza el diseño del gráfico
fig.update_layout(
    title='Importancia de las características en el árbol de decisión',
    xaxis_title='Importancia de las características',
    yaxis_title='Características'
)

# Muestra el gráfico
fig.show()


# Seleccionar las variables relevantes
selected_columns = ['Tenure in Months', 'Monthly Charge', 'Total Regular Charges']

# Crear un nuevo DataFrame con las variables seleccionadas
selected_data = dataset[selected_columns]

# Estandarizar los datos
scaler = StandardScaler()
scaled_data = scaler.fit_transform(selected_data)

# Aplicar PCA
pca = PCA(n_components=2)
pca_data = pca.fit_transform(scaled_data)

# Definir los colores para cada punto
colors = np.where(dataset['Churn Value'] == 0, 'red', 'blue')

# Crear el gráfico de dispersión 2D
fig = go.Figure(data=go.Scatter(
    x=pca_data[:, 0],
    y=pca_data[:, 1],
    mode='markers',
    marker=dict(color=colors, opacity=0.1)
))

# Personalizar el diseño del gráfico
fig.update_layout(
    title='Visualización de datos transformados en dos dimensiones (PCA)',
    xaxis_title='Componente Principal 1',
    yaxis_title='Componente Principal 2'
)

# Mostrar el gráfico
fig.show()


# Seleccionar las variables relevantes
selected_columns = ['Tenure in Months', 'Monthly Charge', 'Total Regular Charges', 'Churn Value']

selected_data = dataset[selected_columns]

# Verificar si hay valores nulos o faltantes en el conjunto de datos
print(selected_data.isnull().sum())

# Visualizar la distribución de las variables numéricas
sns.pairplot(selected_data.select_dtypes(include=['float64']))

# Visualizar la relación entre las variables y el churn
sns.pairplot(selected_data, hue='Churn Value')

Tenure in Months         0
Monthly Charge           0
Total Regular Charges    0
Churn Value              0
dtype: int64

<seaborn.axisgrid.PairGrid at 0x237d36ed820>


# Crear modelo de regresión
reg_model = LinearRegression()

# Entrenar modelo con los datos de entrenamiento
reg_model.fit(X_train, y_train)

# Predecir valores de la variable dependiente utilizando los datos de prueba
y_pred = reg_model.predict(X_test)

# Calcular el R2
r2 = r2_score(y_test, y_pred)

# Imprimir el resultado
print("R2 Score:", r2)

R2 Score: 0.2706898201165513


# Importar la biblioteca scikit-learn


# Generar un conjunto de datos de ejemplo
X, y = make_classification(n_samples=1000, n_features=10, n_classes=2, random_state=42)

# Dividir el conjunto de datos en conjuntos de entrenamiento y prueba
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# Crear una instancia del clasificador KNN
knn = KNeighborsClassifier(n_neighbors=9
                          )

# Entrenar el clasificador con el conjunto de entrenamiento
knn.fit(X_train, y_train)

# Realizar predicciones en el conjunto de prueba
y_pred = knn.predict(X_test)

# Calcular la precisión del clasificador en el conjunto de prueba
accuracy = accuracy_score(y_test, y_pred)

# Imprimir la precisión del clasificador
print("Precisión del clasificador KNN:", accuracy)

Precisión del clasificador KNN: 0.845

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.


# Generar un conjunto de datos de ejemplo
X, y = make_classification(n_samples=1000, n_features=10, n_classes=2, random_state=42)

# Dividir el conjunto de datos en conjuntos de entrenamiento y prueba
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# Definir los valores de k que deseas probar
k_values = range(1, 21)

# Crear una lista para almacenar las precisiones para cada valor de k
accuracies = []

# Iterar sobre los diferentes valores de k
for k in k_values:
    # Crear una instancia del clasificador KNN
    knn = KNeighborsClassifier(n_neighbors=k)

    # Entrenar el clasificador con el conjunto de entrenamiento
    knn.fit(X_train, y_train)

    # Realizar predicciones en el conjunto de prueba
    y_pred = knn.predict(X_test)

    # Calcular la precisión del clasificador en el conjunto de prueba
    accuracy = np.mean(y_pred == y_test)

    # Agregar la precisión a la lista de precisiones
    accuracies.append(accuracy)

# Graficar la precisión en función de los valores de k
plt.plot(k_values, accuracies)
plt.xlabel('k')
plt.ylabel('Precisión')
plt.title('Precisión del clasificador KNN para diferentes valores de k')
plt.show()

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.


# Crear columna Abandono con valores 1 o 0
#dataset['Abandono'] = np.where(dataset['Churn Value'] == 1, 1, 0)
#dataset.dropna(axis=0, how='any', inplace=True)

# Seleccionar las variables relevantes
selected_columns = ['Tenure in Months', 'Monthly Charge', 'Total Regular Charges', 'Total Refunds', 
                    'Total Extra Data Charges', 'Total Long Distance Charges', 'Churn Value']

# Crear un nuevo DataFrame con las variables seleccionadas
selected_data = dataset[selected_columns]

# Estandarizar los datos
scaler = StandardScaler()
scaled_data = scaler.fit_transform(selected_data)

# Definir matriz de características X y vector de destino y
X = scaled_data
y = dataset['Churn Value']

# Crear modelo de clasificación KNN con 9 vecinos
k = 9
modelo = KNeighborsClassifier(n_neighbors=k)

# Entrenar el modelo
modelo.fit(X, y)

KNeighborsClassifier(n_neighbors=9)


# Obtenemos las predicciones del modelo
preds = modelo.predict(X)

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.


preds

array([1, 1, 1, ..., 0, 0, 0], dtype=int64)


(preds == y).sum()

7042


from sklearn.metrics import accuracy_score

# Calcular la exactitud del modelo
accuracy = accuracy_score(y, preds)

print(f"La exactitud del modelo es: {accuracy * 100:.2f}%")

La exactitud del modelo es: 99.99%


from sklearn.metrics import precision_score, recall_score, f1_score

precision = precision_score(y, preds)
recall = recall_score(y, preds)
f1 = f1_score(y, preds)

print(f"Precision: {precision * 100:.2f}%")
print(f"Recall: {recall * 100:.2f}%")
print(f"F1-score: {f1 * 100:.2f}%")

Precision: 100.00%
Recall: 99.95%
F1-score: 99.97%


n_samples, n_features = X.shape

print(f"Se usaron {n_samples} datos para entrenar el modelo.")

Se usaron 7043 datos para entrenar el modelo.


# Separamos en entrenamiento/test en razon 70/30 %
#0-100  0-1
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=1, shuffle=False)


X_train.shape

(4930, 7)


k = 9
modelo = KNeighborsClassifier(n_neighbors=k)

# Entrenar el modelo
modelo.fit(X_train, y_train)

KNeighborsClassifier(n_neighbors=9)


preds_test = modelo.predict(X_test)

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.


(preds_test == y_test).sum()

2112


(preds_test == y_test).sum()/y.sum()*100

113.0016051364366


accuracy_test = accuracy_score(y_test, preds_test)

print(f"La exactitud del modelo en el conjunto de prueba es: {accuracy_test * 100:.2f}%")

La exactitud del modelo en el conjunto de prueba es: 99.95%


# Predecir con los datos de prueba
preds_test = modelo.predict(X_test)

# Calcular la exactitud del modelo en el conjunto de prueba
accuracy_test = accuracy_score(y_test, preds_test)

# Calcular la precisión, el recall y el F1-score en el conjunto de prueba
precision_test = precision_score(y_test, preds_test)
recall_test = recall_score(y_test, preds_test)
f1_test = f1_score(y_test, preds_test)

print(f"Accuracy: {accuracy_test * 100:.2f}%")
print(f"Precision: {precision_test * 100:.2f}%")
print(f"Recall: {recall_test * 100:.2f}%")
print(f"F1-score: {f1_test * 100:.2f}%")

Accuracy: 99.95%
Precision: 0.00%
Recall: 0.00%
F1-score: 0.00%

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\neighbors\_classification.py:228: FutureWarning:

Unlike other reduction functions (e.g. `skew`, `kurtosis`), the default behavior of `mode` typically preserves the axis it acts along. In SciPy 1.11.0, this behavior will change: the default value of `keepdims` will become False, the `axis` over which the statistic is taken will be eliminated, and the value None will no longer be accepted. Set `keepdims` to True or False to avoid this warning.

C:\Users\jtroc\anaconda3\lib\site-packages\sklearn\metrics\_classification.py:1318: UndefinedMetricWarning:

Recall is ill-defined and being set to 0.0 due to no true samples. Use `zero_division` parameter to control this behavior.


# Calcular y visualizar la matriz de confusión
cm = confusion_matrix(y_test, preds_test)
sns.heatmap(cm, annot=True, fmt='d', cmap='Blues')
plt.ylabel('True label')
plt.xlabel('Predicted label')
plt.show()


plt.figure(figsize=(10, 6))
cm = confusion_matrix(y_test, preds_test, labels=modelo.classes_)
disp = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=modelo.classes_)
disp.plot()
plt.show()

<Figure size 1000x600 with 0 Axes>


from sklearn.svm import SVC
from sklearn.metrics import classification_report, confusion_matrix
import matplotlib.pyplot as plt
import seaborn as sns

# Crear un objeto de la clase SVC con un núcleo específico (por ejemplo, 'rbf')
modelo = SVC(kernel='rbf')

# Entrenar el modelo usando los datos de entrenamiento (X_train, y_train)
modelo.fit(X_train, y_train)

# Predecir las etiquetas de clase para los datos de prueba (X_test)
predicciones = modelo.predict(X_test)

# Matriz de confusión
matriz_confusion = confusion_matrix(y_test, predicciones)
print("Matriz de confusión:")
print(matriz_confusion)

# Informe de clasificación
reporte_clasificacion = classification_report(y_test, predicciones)
print("Informe de clasificación:")
print(reporte_clasificacion)

# Visualización de la matriz de confusión
sns.heatmap(matriz_confusion, annot=True, cmap='Blues')
plt.xlabel('Etiqueta Predicha')
plt.ylabel('Etiqueta Real')
plt.title('Matriz de Confusión')
plt.show()

Matriz de confusión:
[[2113]]
Informe de clasificación:
              precision    recall  f1-score   support

           0       1.00      1.00      1.00      2113

    accuracy                           1.00      2113
   macro avg       1.00      1.00      1.00      2113
weighted avg       1.00      1.00      1.00      2113


# Selecciona las columnas que deseas incluir en el clustering
feature_columns = ['Tenure in Months', 'Monthly Charge', 'Total Regular Charges', 'Total Refunds', 'Total Extra Data Charges', 'Total Long Distance Charges']

# Escala tus datos para que tengan media cero y varianza uno (opcional pero recomendado)
scaler = StandardScaler()
X = scaler.fit_transform(dataset[feature_columns])

# Crea una lista de valores de k para iterar
k_values = range(2, 10)

# Itera sobre los diferentes valores de k y almacena la puntuación de Silhouette en una lista
silhouette_scores = []
for k in k_values:
    # Crea el modelo de K-means con el valor de k actual
    kmeans = KMeans(n_clusters=k,n_init=50, random_state=42)
    kmeans.fit(X)
    
    # Calcula la puntuación de Silhouette del modelo actual y agrega a una lista
    silhouette_score_kmeans = silhouette_score(X, kmeans.labels_)
    silhouette_scores.append(silhouette_score_kmeans)
    
    # Imprime la puntuación de Silhouette del modelo actual
    print("K-means silhouette score for k =", k, ": ", silhouette_score_kmeans)

# Encuentra el mejor valor de k
best_k = k_values[np.argmax(silhouette_scores)]
print("Best k value based on silhouette score:", best_k)

# Visualiza los valores de la puntuación de Silhouette para cada valor de k
fig, ax = plt.subplots(figsize=(12, 8))
ax.plot(k_values, silhouette_scores, marker='o')
ax.set_xlabel('Número de clusters (k)')
ax.set_ylabel('Puntuación de Silhouette')
ax.set_title('Puntuación de Silhouette para diferentes valores de k')
plt.show()

K-means silhouette score for k = 2 :  0.39533396161100065
K-means silhouette score for k = 3 :  0.42367875578062864
K-means silhouette score for k = 4 :  0.4010040681701614
K-means silhouette score for k = 5 :  0.36107230176841454
K-means silhouette score for k = 6 :  0.349104619325899
K-means silhouette score for k = 7 :  0.3408311699962871
K-means silhouette score for k = 8 :  0.3415226726544426
K-means silhouette score for k = 9 :  0.3366729745185902
Best k value based on silhouette score: 3


# Visualiza el árbol de decisión
plt.figure(figsize=(20, 10))
plot_tree(clf, feature_names=feature_columns, class_names=['No Churn', 'Churn'], filled=True)
plt.show()


# Calcular la matriz de correlación
correlation_matrix = dataset.corr()

# Crear un mapa de calor utilizando Seaborn
plt.figure(figsize=(20, 20))
sns.heatmap(correlation_matrix, annot=True, cmap="coolwarm", linewidths=0.5, square=True)

plt.title("Mapa de correlación de todas las variables")
plt.show()

	Customer ID	Referred a Friend	Number of Referrals	Tenure in Months	Offer	Phone Service	Avg Monthly Long Distance Charges	Multiple Lines	Internet Service	Internet Type	Avg Monthly GB Download	Online Security	Online Backup	Device Protection Plan	Premium Tech Support	Streaming TV	Streaming Movies	Streaming Music	Unlimited Data	Contract	Paperless Billing	Payment Method	Monthly Charge	Total Regular Charges	Total Refunds	Total Extra Data Charges	Total Long Distance Charges	Gender	Age	Under 30	Senior Citizen	Married	Dependents	Number of Dependents	City	Zip Code	Latitude	Longitude	Population	Churn Value	CLTV	Churn Category	Churn Reason	Total Customer Svc Requests	Product/Service Issues Reported	Customer Satisfaction
0	8779-QRDMV	No	0	1	None	No	0.00	No	Yes	Fiber Optic	9	No	No	Yes	No	No	Yes	No	No	Month-to-Month	Yes	Bank Withdrawal	41.236	39.65	0.00	0.0	0.00	Male	78	No	Yes	No	No	0	Los Angeles	90022	34.023810	-118.156582	68701	1	5433	Competitor	Competitor offered more data	5	0	NaN
1	7495-OOKFY	Yes	1	8	Offer E	Yes	48.85	Yes	Yes	Cable	19	No	Yes	No	No	No	No	No	No	Month-to-Month	Yes	Credit Card	83.876	633.30	0.00	120.0	390.80	Female	74	No	Yes	Yes	Yes	1	Los Angeles	90063	34.044271	-118.185237	55668	1	5302	Competitor	Competitor made better offer	5	0	NaN
2	1658-BYGOY	No	0	18	Offer D	Yes	11.33	Yes	Yes	Fiber Optic	57	No	No	No	No	Yes	Yes	Yes	Yes	Month-to-Month	Yes	Bank Withdrawal	99.268	1752.55	45.61	0.0	203.94	Male	71	No	Yes	No	Yes	3	Los Angeles	90065	34.108833	-118.229715	47534	1	3179	Competitor	Competitor made better offer	1	0	NaN
3	4598-XLKNJ	Yes	1	25	Offer C	Yes	19.76	No	Yes	Fiber Optic	13	No	Yes	Yes	No	Yes	Yes	No	No	Month-to-Month	Yes	Bank Withdrawal	102.440	2514.50	13.43	327.0	494.00	Female	78	No	Yes	Yes	Yes	1	Inglewood	90303	33.936291	-118.332639	27778	1	5337	Dissatisfaction	Limited range of services	1	1	2.0
4	4846-WHAFZ	Yes	1	37	Offer C	Yes	6.33	Yes	Yes	Cable	15	No	No	No	No	No	No	No	No	Month-to-Month	Yes	Bank Withdrawal	79.560	2868.15	0.00	430.0	234.21	Female	80	No	Yes	Yes	Yes	1	Whittier	90602	33.972119	-118.020188	26265	1	2793	Price	Extra data charges	1	0	2.0

	Number of Referrals	Tenure in Months	Avg Monthly Long Distance Charges	Avg Monthly GB Download	Monthly Charge	Total Regular Charges	Total Refunds	Total Extra Data Charges	Total Long Distance Charges	Age	Number of Dependents	Zip Code	Latitude	Longitude	Population	Churn Value	CLTV	Total Customer Svc Requests	Product/Service Issues Reported	Customer Satisfaction
count	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	7043.000000	1834.000000
mean	1.951867	32.386767	22.958954	21.110890	65.538800	2280.381264	1.962182	278.499225	749.099262	46.509726	0.468692	93486.070567	36.197455	-119.756684	22139.603294	0.265370	4400.295755	1.338776	0.308107	3.005453
std	3.001199	24.542061	15.448113	20.948471	30.606805	2266.220462	7.902614	685.039625	846.660055	16.750352	0.962802	1856.767505	2.468929	2.154425	21152.392837	0.441561	1183.057152	1.430471	0.717514	1.256938
min	0.000000	1.000000	0.000000	0.000000	18.250000	18.800000	0.000000	0.000000	0.000000	19.000000	0.000000	90001.000000	32.555828	-124.301372	11.000000	0.000000	2003.000000	0.000000	0.000000	1.000000
25%	0.000000	9.000000	9.210000	3.000000	35.890000	400.150000	0.000000	0.000000	70.545000	32.000000	0.000000	92101.000000	33.990646	-121.788090	2344.000000	0.000000	3469.000000	0.000000	0.000000	2.000000
50%	0.000000	29.000000	22.890000	17.000000	71.968000	1394.550000	0.000000	0.000000	401.440000	46.000000	0.000000	93518.000000	36.205465	-119.595293	17554.000000	0.000000	4527.000000	1.000000	0.000000	3.000000
75%	3.000000	55.000000	36.395000	28.000000	90.650000	3786.600000	0.000000	182.620000	1191.100000	60.000000	0.000000	95329.000000	38.161321	-117.969795	36125.000000	1.000000	5380.500000	2.000000	0.000000	4.000000
max	11.000000	72.000000	49.990000	94.000000	123.084000	8684.800000	49.790000	6477.000000	3564.720000	80.000000	9.000000	96150.000000	41.962127	-114.192901	105285.000000	1.000000	6500.000000	9.000000	6.000000	5.000000

	Conteo	Porcentaje (%)
1.0	332.0	18.10
2.0	200.0	10.91
3.0	675.0	36.80
4.0	380.0	20.72
5.0	247.0	13.47
Total	1834.0	100.00

Analisis de abandono de clientes¶

Descripcion del proyecto¶

Descripcion del Dataset¶

Importacion de librerias utilizadas¶

Importacion del data set¶

Muestreo de datos de todas las columnas:¶

Informacion referente a las columnas¶

Descripcion de las columnas:¶

Resumen Estadístico del Conjunto de Datos¶

Visualizacion del abandono¶

Visualizacion de la satifaccion del cliente¶

Resumen¶

¿Existe una relación entre la satisfacción del cliente y la duración de la suscripción?¶

¿Existe una relación entre la tasa de cancelación y la satisfacción del cliente?¶

¿Influye el tiempo de permanencia en la tasa de cancelación?¶

Análisis de los Factores que Afectan la Satisfacción del Cliente¶

¿Influyen los cargos mensuales en la satisfacción del cliente y la tasa de cancelación?¶

¿Existe una relación entre el número de solicitudes de servicio al cliente y la satisfacción del cliente?¶

¿Existe una relación entre la cantidad de problemas de producto/servicio reportados y la satisfacción del cliente?¶

Mapa de Correlación de Variables Seleccionadas¶

API¶

Predicción del abandono de clientes (Churn Value)¶